BACILLUS WELCH 11 531 
2 to 1.' Wolf and Harris find that the volume of gas produced in carbo- 
hydrate media (lactose 3 to 4 per cent) may be nearly four times that 
of the original amount of medium. Neill and Fleming- have obtained 
carbohydrases aufl a lipase from B. welchii. Nitrogen has not been 
definitely detected in the fermentation gases.'' 
About 60 per cent of the acids formed by B. welchii are volatile, 
principally butyric. Tissier and IMartelly state that propionic and 
acetic acids are also found in the volatile fraction. Lactic acid is the 
principal non-volatile acid product. Taylor^ found that 10 cc. of a 
glucose broth culture of B. welchii generated enough gas in a few hours 
to raise the needle of a pressure gauge to the amazing equivalent of 
23-pounds' pressure (H atmospheres). He believes the gas formed 
by the organism in the bloodvessels as a result of infection is in itself 
not toxic, but of great importance mechanically, since it causes death 
of the surrounding tissues by pressure, the actual fragmentation 
of tissues, especially muscles, and thereby a mechanical scattering of 
infection to more distant areas. A histamine-like substance is found 
in some cultures of the gas bacillus.^ 
Toxin Formation. — Vassm'i^ appears to have been the first to dis- 
cover toxic substances in filtrates from cultures of B. welchii. These 
were found to be thermostabile, but possessed no antigenic properties. 
Brown ^ states that a toxin forms in sugar-free broth which, when freed 
from bacilli, is pathogenic for guinea-pigs. The acids produced during 
the growth of the organism are said to be poisonous for experimental 
animals;* when they are neutralized the poisons become inert. Bull 
and Pritchett,^ however have shown that filtrates of cultures grown 
in acid media (3 per cent acid to phenolphthalein) contain a soluble, 
poisonous substance which reproduces the hemolytic and necrotic 
changes in pigeons that are peculiar to B. welchii. The poisonous 
substance is thermostabile and is destroyed by heating to 70° C. for 
an hour. Henry"' has studied the toxin in greater detail, and finds that 
it may be fractionated into a hemotoxin, to which mice are very sus- 
ceptible, and a myotoxin, to which rabbits are susceptible. Injections 
into large animals lead to the formation of substances in the serum that 
protect experimental laboratory animals against multiple fatal doses of 
the poison." 
I Simonds: Loc. cit., p. 40. - Jour. Exp. Med., 1927, 45, 947. 
» Dunham (Bull. Johns Hopkins Hosp., 1897, 8, 68), however, states that as much as 
8 per cent of nitrogen may be present. This is unconfirmed up to the present time. 
* Lancet. 1916, i, 12.3. 
6 Kendall and Schmitt: Jour. Infec. Dis., 1926, 39, 250. Kendall: Ibid., 1927, 
40, 689. 
« Wien. klin. Wchnschr., 1905, 18, 921. 
' Ann. Report of Massachusetts State Board of Health, 1909. 
8 McCampbell: Jour. Infec. Dis., 1909, 6, 5.37. Stewart and West: Jour. Immunol., 
1916, 1, 189. 
5 .Jour. Exp. Med., 1917, 26, 119. DeKruif, Adams and Ireland: Jour. Infec. Dis., 
1917, 21, 580. 
i» Jour. Pathol, and Bacteriol., 1923, 26, 497. 
II Bull and Pritchett: Jour. Exp. Med., 1917, 26, 603, 867. 
